llvm/test/CodeGen/AMDGPU/promote-alloca-vector-gep.ll

; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 5
; RUN: opt -S -mtriple=amdgcn-amd-amdhsa -passes=amdgpu-promote-alloca < %s | FileCheck %s

; Check that invalid IR is not produced on a vector typed
; getelementptr with a scalar alloca pointer base.

define amdgpu_kernel void @scalar_alloca_ptr_with_vector_gep_offset() {
; CHECK-LABEL: define amdgpu_kernel void @scalar_alloca_ptr_with_vector_gep_offset() {
; CHECK-NEXT:  [[BB:.*:]]
; CHECK-NEXT:    [[TMP0:%.*]] = call noalias nonnull dereferenceable(64) ptr addrspace(4) @llvm.amdgcn.dispatch.ptr()
; CHECK-NEXT:    [[TMP1:%.*]] = getelementptr inbounds i32, ptr addrspace(4) [[TMP0]], i64 1
; CHECK-NEXT:    [[TMP2:%.*]] = load i32, ptr addrspace(4) [[TMP1]], align 4, !invariant.load [[META0:![0-9]+]]
; CHECK-NEXT:    [[TMP3:%.*]] = getelementptr inbounds i32, ptr addrspace(4) [[TMP0]], i64 2
; CHECK-NEXT:    [[TMP4:%.*]] = load i32, ptr addrspace(4) [[TMP3]], align 4, !range [[RNG1:![0-9]+]], !invariant.load [[META0]]
; CHECK-NEXT:    [[TMP5:%.*]] = lshr i32 [[TMP2]], 16
; CHECK-NEXT:    [[TMP6:%.*]] = call range(i32 0, 1024) i32 @llvm.amdgcn.workitem.id.x()
; CHECK-NEXT:    [[TMP7:%.*]] = call range(i32 0, 1024) i32 @llvm.amdgcn.workitem.id.y()
; CHECK-NEXT:    [[TMP8:%.*]] = call range(i32 0, 1024) i32 @llvm.amdgcn.workitem.id.z()
; CHECK-NEXT:    [[TMP9:%.*]] = mul nuw nsw i32 [[TMP5]], [[TMP4]]
; CHECK-NEXT:    [[TMP10:%.*]] = mul i32 [[TMP9]], [[TMP6]]
; CHECK-NEXT:    [[TMP11:%.*]] = mul nuw nsw i32 [[TMP7]], [[TMP4]]
; CHECK-NEXT:    [[TMP12:%.*]] = add i32 [[TMP10]], [[TMP11]]
; CHECK-NEXT:    [[TMP13:%.*]] = add i32 [[TMP12]], [[TMP8]]
; CHECK-NEXT:    [[TMP14:%.*]] = getelementptr inbounds [1024 x [4 x i32]], ptr addrspace(3) @scalar_alloca_ptr_with_vector_gep_offset.alloca, i32 0, i32 [[TMP13]]
; CHECK-NEXT:    [[GETELEMENTPTR:%.*]] = getelementptr inbounds i8, ptr addrspace(3) [[TMP14]], <4 x i64> <i64 0, i64 1, i64 2, i64 3>
; CHECK-NEXT:    [[EXTRACTELEMENT:%.*]] = extractelement <4 x ptr addrspace(3)> [[GETELEMENTPTR]], i64 0
; CHECK-NEXT:    store i32 0, ptr addrspace(3) [[EXTRACTELEMENT]], align 4
; CHECK-NEXT:    ret void
;
bb:
  %alloca = alloca [4 x i32], align 4, addrspace(5)
  %getelementptr = getelementptr inbounds i8, ptr addrspace(5) %alloca, <4 x i64> <i64 0, i64 1, i64 2, i64 3>
  %extractelement = extractelement <4 x ptr addrspace(5)> %getelementptr, i64 0
  store i32 0, ptr addrspace(5) %extractelement
  ret void
}

define amdgpu_kernel void @scalar_alloca_ptr_with_vector_gep_offset_select(i1 %cond) {
; CHECK-LABEL: define amdgpu_kernel void @scalar_alloca_ptr_with_vector_gep_offset_select(
; CHECK-SAME: i1 [[COND:%.*]]) {
; CHECK-NEXT:  [[BB:.*:]]
; CHECK-NEXT:    [[TMP0:%.*]] = call noalias nonnull dereferenceable(64) ptr addrspace(4) @llvm.amdgcn.dispatch.ptr()
; CHECK-NEXT:    [[TMP1:%.*]] = getelementptr inbounds i32, ptr addrspace(4) [[TMP0]], i64 1
; CHECK-NEXT:    [[TMP2:%.*]] = load i32, ptr addrspace(4) [[TMP1]], align 4, !invariant.load [[META0]]
; CHECK-NEXT:    [[TMP3:%.*]] = getelementptr inbounds i32, ptr addrspace(4) [[TMP0]], i64 2
; CHECK-NEXT:    [[TMP4:%.*]] = load i32, ptr addrspace(4) [[TMP3]], align 4, !range [[RNG1]], !invariant.load [[META0]]
; CHECK-NEXT:    [[TMP5:%.*]] = lshr i32 [[TMP2]], 16
; CHECK-NEXT:    [[TMP6:%.*]] = call range(i32 0, 1024) i32 @llvm.amdgcn.workitem.id.x()
; CHECK-NEXT:    [[TMP7:%.*]] = call range(i32 0, 1024) i32 @llvm.amdgcn.workitem.id.y()
; CHECK-NEXT:    [[TMP8:%.*]] = call range(i32 0, 1024) i32 @llvm.amdgcn.workitem.id.z()
; CHECK-NEXT:    [[TMP9:%.*]] = mul nuw nsw i32 [[TMP5]], [[TMP4]]
; CHECK-NEXT:    [[TMP10:%.*]] = mul i32 [[TMP9]], [[TMP6]]
; CHECK-NEXT:    [[TMP11:%.*]] = mul nuw nsw i32 [[TMP7]], [[TMP4]]
; CHECK-NEXT:    [[TMP12:%.*]] = add i32 [[TMP10]], [[TMP11]]
; CHECK-NEXT:    [[TMP13:%.*]] = add i32 [[TMP12]], [[TMP8]]
; CHECK-NEXT:    [[TMP14:%.*]] = getelementptr inbounds [1024 x [4 x i32]], ptr addrspace(3) @scalar_alloca_ptr_with_vector_gep_offset_select.alloca, i32 0, i32 [[TMP13]]
; CHECK-NEXT:    [[GETELEMENTPTR0:%.*]] = getelementptr inbounds i8, ptr addrspace(3) [[TMP14]], <4 x i64> <i64 0, i64 1, i64 2, i64 3>
; CHECK-NEXT:    [[GETELEMENTPTR1:%.*]] = getelementptr inbounds i8, ptr addrspace(3) [[TMP14]], <4 x i64> <i64 3, i64 2, i64 1, i64 0>
; CHECK-NEXT:    [[SELECT:%.*]] = select i1 [[COND]], <4 x ptr addrspace(3)> [[GETELEMENTPTR0]], <4 x ptr addrspace(3)> [[GETELEMENTPTR1]]
; CHECK-NEXT:    [[EXTRACTELEMENT:%.*]] = extractelement <4 x ptr addrspace(3)> [[SELECT]], i64 1
; CHECK-NEXT:    store i32 0, ptr addrspace(3) [[EXTRACTELEMENT]], align 4
; CHECK-NEXT:    ret void
;
bb:
  %alloca = alloca [4 x i32], align 4, addrspace(5)
  %getelementptr0 = getelementptr inbounds i8, ptr addrspace(5) %alloca, <4 x i64> <i64 0, i64 1, i64 2, i64 3>
  %getelementptr1 = getelementptr inbounds i8, ptr addrspace(5) %alloca, <4 x i64> <i64 3, i64 2, i64 1, i64 0>
  %select = select i1 %cond, <4 x ptr addrspace(5)> %getelementptr0, <4 x ptr addrspace(5)> %getelementptr1
  %extractelement = extractelement <4 x ptr addrspace(5)> %select, i64 1
  store i32 0, ptr addrspace(5) %extractelement
  ret void
}

define amdgpu_kernel void @scalar_alloca_ptr_with_vector_gep_offset_select_nullptr0(i1 %cond) {
; CHECK-LABEL: define amdgpu_kernel void @scalar_alloca_ptr_with_vector_gep_offset_select_nullptr0(
; CHECK-SAME: i1 [[COND:%.*]]) {
; CHECK-NEXT:  [[BB:.*:]]
; CHECK-NEXT:    [[TMP0:%.*]] = call noalias nonnull dereferenceable(64) ptr addrspace(4) @llvm.amdgcn.dispatch.ptr()
; CHECK-NEXT:    [[TMP1:%.*]] = getelementptr inbounds i32, ptr addrspace(4) [[TMP0]], i64 1
; CHECK-NEXT:    [[TMP2:%.*]] = load i32, ptr addrspace(4) [[TMP1]], align 4, !invariant.load [[META0]]
; CHECK-NEXT:    [[TMP3:%.*]] = getelementptr inbounds i32, ptr addrspace(4) [[TMP0]], i64 2
; CHECK-NEXT:    [[TMP4:%.*]] = load i32, ptr addrspace(4) [[TMP3]], align 4, !range [[RNG1]], !invariant.load [[META0]]
; CHECK-NEXT:    [[TMP5:%.*]] = lshr i32 [[TMP2]], 16
; CHECK-NEXT:    [[TMP6:%.*]] = call range(i32 0, 1024) i32 @llvm.amdgcn.workitem.id.x()
; CHECK-NEXT:    [[TMP7:%.*]] = call range(i32 0, 1024) i32 @llvm.amdgcn.workitem.id.y()
; CHECK-NEXT:    [[TMP8:%.*]] = call range(i32 0, 1024) i32 @llvm.amdgcn.workitem.id.z()
; CHECK-NEXT:    [[TMP9:%.*]] = mul nuw nsw i32 [[TMP5]], [[TMP4]]
; CHECK-NEXT:    [[TMP10:%.*]] = mul i32 [[TMP9]], [[TMP6]]
; CHECK-NEXT:    [[TMP11:%.*]] = mul nuw nsw i32 [[TMP7]], [[TMP4]]
; CHECK-NEXT:    [[TMP12:%.*]] = add i32 [[TMP10]], [[TMP11]]
; CHECK-NEXT:    [[TMP13:%.*]] = add i32 [[TMP12]], [[TMP8]]
; CHECK-NEXT:    [[TMP14:%.*]] = getelementptr inbounds [1024 x [4 x i32]], ptr addrspace(3) @scalar_alloca_ptr_with_vector_gep_offset_select_nullptr0.alloca, i32 0, i32 [[TMP13]]
; CHECK-NEXT:    [[GETELEMENTPTR0:%.*]] = getelementptr inbounds i8, ptr addrspace(3) [[TMP14]], <4 x i64> <i64 0, i64 1, i64 2, i64 3>
; CHECK-NEXT:    [[SELECT:%.*]] = select i1 [[COND]], <4 x ptr addrspace(3)> zeroinitializer, <4 x ptr addrspace(3)> [[GETELEMENTPTR0]]
; CHECK-NEXT:    [[EXTRACTELEMENT:%.*]] = extractelement <4 x ptr addrspace(3)> [[SELECT]], i64 1
; CHECK-NEXT:    store i32 0, ptr addrspace(3) [[EXTRACTELEMENT]], align 4
; CHECK-NEXT:    ret void
;
bb:
  %alloca = alloca [4 x i32], align 4, addrspace(5)
  %getelementptr0 = getelementptr inbounds i8, ptr addrspace(5) %alloca, <4 x i64> <i64 0, i64 1, i64 2, i64 3>
  %select = select i1 %cond, <4 x ptr addrspace(5)> zeroinitializer, <4 x ptr addrspace(5)> %getelementptr0
  %extractelement = extractelement <4 x ptr addrspace(5)> %select, i64 1
  store i32 0, ptr addrspace(5) %extractelement
  ret void
}

define amdgpu_kernel void @scalar_alloca_ptr_with_vector_gep_offset_select_nullptr1(i1 %cond) {
; CHECK-LABEL: define amdgpu_kernel void @scalar_alloca_ptr_with_vector_gep_offset_select_nullptr1(
; CHECK-SAME: i1 [[COND:%.*]]) {
; CHECK-NEXT:  [[BB:.*:]]
; CHECK-NEXT:    [[TMP0:%.*]] = call noalias nonnull dereferenceable(64) ptr addrspace(4) @llvm.amdgcn.dispatch.ptr()
; CHECK-NEXT:    [[TMP1:%.*]] = getelementptr inbounds i32, ptr addrspace(4) [[TMP0]], i64 1
; CHECK-NEXT:    [[TMP2:%.*]] = load i32, ptr addrspace(4) [[TMP1]], align 4, !invariant.load [[META0]]
; CHECK-NEXT:    [[TMP3:%.*]] = getelementptr inbounds i32, ptr addrspace(4) [[TMP0]], i64 2
; CHECK-NEXT:    [[TMP4:%.*]] = load i32, ptr addrspace(4) [[TMP3]], align 4, !range [[RNG1]], !invariant.load [[META0]]
; CHECK-NEXT:    [[TMP5:%.*]] = lshr i32 [[TMP2]], 16
; CHECK-NEXT:    [[TMP6:%.*]] = call range(i32 0, 1024) i32 @llvm.amdgcn.workitem.id.x()
; CHECK-NEXT:    [[TMP7:%.*]] = call range(i32 0, 1024) i32 @llvm.amdgcn.workitem.id.y()
; CHECK-NEXT:    [[TMP8:%.*]] = call range(i32 0, 1024) i32 @llvm.amdgcn.workitem.id.z()
; CHECK-NEXT:    [[TMP9:%.*]] = mul nuw nsw i32 [[TMP5]], [[TMP4]]
; CHECK-NEXT:    [[TMP10:%.*]] = mul i32 [[TMP9]], [[TMP6]]
; CHECK-NEXT:    [[TMP11:%.*]] = mul nuw nsw i32 [[TMP7]], [[TMP4]]
; CHECK-NEXT:    [[TMP12:%.*]] = add i32 [[TMP10]], [[TMP11]]
; CHECK-NEXT:    [[TMP13:%.*]] = add i32 [[TMP12]], [[TMP8]]
; CHECK-NEXT:    [[TMP14:%.*]] = getelementptr inbounds [1024 x [4 x i32]], ptr addrspace(3) @scalar_alloca_ptr_with_vector_gep_offset_select_nullptr1.alloca, i32 0, i32 [[TMP13]]
; CHECK-NEXT:    [[GETELEMENTPTR0:%.*]] = getelementptr inbounds i8, ptr addrspace(3) [[TMP14]], <4 x i64> <i64 0, i64 1, i64 2, i64 3>
; CHECK-NEXT:    [[SELECT:%.*]] = select i1 [[COND]], <4 x ptr addrspace(3)> [[GETELEMENTPTR0]], <4 x ptr addrspace(3)> zeroinitializer
; CHECK-NEXT:    [[EXTRACTELEMENT:%.*]] = extractelement <4 x ptr addrspace(3)> [[SELECT]], i64 1
; CHECK-NEXT:    store i32 0, ptr addrspace(3) [[EXTRACTELEMENT]], align 4
; CHECK-NEXT:    ret void
;
bb:
  %alloca = alloca [4 x i32], align 4, addrspace(5)
  %getelementptr0 = getelementptr inbounds i8, ptr addrspace(5) %alloca, <4 x i64> <i64 0, i64 1, i64 2, i64 3>
  %select = select i1 %cond, <4 x ptr addrspace(5)> %getelementptr0, <4 x ptr addrspace(5)> zeroinitializer
  %extractelement = extractelement <4 x ptr addrspace(5)> %select, i64 1
  store i32 0, ptr addrspace(5) %extractelement
  ret void
}

define amdgpu_kernel void @scalar_alloca_ptr_with_vector_gep_offset_icmp_nullptr0(i1 %cond, ptr addrspace(1) %out) {
; CHECK-LABEL: define amdgpu_kernel void @scalar_alloca_ptr_with_vector_gep_offset_icmp_nullptr0(
; CHECK-SAME: i1 [[COND:%.*]], ptr addrspace(1) [[OUT:%.*]]) {
; CHECK-NEXT:  [[BB0:.*:]]
; CHECK-NEXT:    [[TMP0:%.*]] = call noalias nonnull dereferenceable(64) ptr addrspace(4) @llvm.amdgcn.dispatch.ptr()
; CHECK-NEXT:    [[TMP1:%.*]] = getelementptr inbounds i32, ptr addrspace(4) [[TMP0]], i64 1
; CHECK-NEXT:    [[TMP2:%.*]] = load i32, ptr addrspace(4) [[TMP1]], align 4, !invariant.load [[META0]]
; CHECK-NEXT:    [[TMP3:%.*]] = getelementptr inbounds i32, ptr addrspace(4) [[TMP0]], i64 2
; CHECK-NEXT:    [[TMP4:%.*]] = load i32, ptr addrspace(4) [[TMP3]], align 4, !range [[RNG1]], !invariant.load [[META0]]
; CHECK-NEXT:    [[TMP5:%.*]] = lshr i32 [[TMP2]], 16
; CHECK-NEXT:    [[TMP6:%.*]] = call range(i32 0, 1024) i32 @llvm.amdgcn.workitem.id.x()
; CHECK-NEXT:    [[TMP7:%.*]] = call range(i32 0, 1024) i32 @llvm.amdgcn.workitem.id.y()
; CHECK-NEXT:    [[TMP8:%.*]] = call range(i32 0, 1024) i32 @llvm.amdgcn.workitem.id.z()
; CHECK-NEXT:    [[TMP9:%.*]] = mul nuw nsw i32 [[TMP5]], [[TMP4]]
; CHECK-NEXT:    [[TMP10:%.*]] = mul i32 [[TMP9]], [[TMP6]]
; CHECK-NEXT:    [[TMP11:%.*]] = mul nuw nsw i32 [[TMP7]], [[TMP4]]
; CHECK-NEXT:    [[TMP12:%.*]] = add i32 [[TMP10]], [[TMP11]]
; CHECK-NEXT:    [[TMP13:%.*]] = add i32 [[TMP12]], [[TMP8]]
; CHECK-NEXT:    [[TMP14:%.*]] = getelementptr inbounds [1024 x [4 x i32]], ptr addrspace(3) @scalar_alloca_ptr_with_vector_gep_offset_icmp_nullptr0.alloca, i32 0, i32 [[TMP13]]
; CHECK-NEXT:    [[GETELEMENTPTR0:%.*]] = getelementptr inbounds i8, ptr addrspace(3) [[TMP14]], <4 x i64> <i64 0, i64 1, i64 2, i64 3>
; CHECK-NEXT:    [[SELECT:%.*]] = select i1 [[COND]], <4 x ptr addrspace(3)> zeroinitializer, <4 x ptr addrspace(3)> [[GETELEMENTPTR0]]
; CHECK-NEXT:    [[ICMP:%.*]] = icmp eq <4 x ptr addrspace(3)> [[SELECT]], zeroinitializer
; CHECK-NEXT:    store <4 x i1> [[ICMP]], ptr addrspace(1) [[OUT]], align 1
; CHECK-NEXT:    ret void
;
bb0:
  %alloca = alloca [4 x i32], align 4, addrspace(5)
  %getelementptr0 = getelementptr inbounds i8, ptr addrspace(5) %alloca, <4 x i64> <i64 0, i64 1, i64 2, i64 3>
  %select = select i1 %cond, <4 x ptr addrspace(5)> zeroinitializer, <4 x ptr addrspace(5)> %getelementptr0
  %icmp = icmp eq <4 x ptr addrspace(5)> %select, zeroinitializer
  store <4 x i1> %icmp, ptr addrspace(1) %out
  ret void
}

define amdgpu_kernel void @scalar_alloca_ptr_with_vector_gep_offset_icmp_nullptr1(i1 %cond, ptr addrspace(1) %out) {
; CHECK-LABEL: define amdgpu_kernel void @scalar_alloca_ptr_with_vector_gep_offset_icmp_nullptr1(
; CHECK-SAME: i1 [[COND:%.*]], ptr addrspace(1) [[OUT:%.*]]) {
; CHECK-NEXT:  [[BB0:.*:]]
; CHECK-NEXT:    [[TMP0:%.*]] = call noalias nonnull dereferenceable(64) ptr addrspace(4) @llvm.amdgcn.dispatch.ptr()
; CHECK-NEXT:    [[TMP1:%.*]] = getelementptr inbounds i32, ptr addrspace(4) [[TMP0]], i64 1
; CHECK-NEXT:    [[TMP2:%.*]] = load i32, ptr addrspace(4) [[TMP1]], align 4, !invariant.load [[META0]]
; CHECK-NEXT:    [[TMP3:%.*]] = getelementptr inbounds i32, ptr addrspace(4) [[TMP0]], i64 2
; CHECK-NEXT:    [[TMP4:%.*]] = load i32, ptr addrspace(4) [[TMP3]], align 4, !range [[RNG1]], !invariant.load [[META0]]
; CHECK-NEXT:    [[TMP5:%.*]] = lshr i32 [[TMP2]], 16
; CHECK-NEXT:    [[TMP6:%.*]] = call range(i32 0, 1024) i32 @llvm.amdgcn.workitem.id.x()
; CHECK-NEXT:    [[TMP7:%.*]] = call range(i32 0, 1024) i32 @llvm.amdgcn.workitem.id.y()
; CHECK-NEXT:    [[TMP8:%.*]] = call range(i32 0, 1024) i32 @llvm.amdgcn.workitem.id.z()
; CHECK-NEXT:    [[TMP9:%.*]] = mul nuw nsw i32 [[TMP5]], [[TMP4]]
; CHECK-NEXT:    [[TMP10:%.*]] = mul i32 [[TMP9]], [[TMP6]]
; CHECK-NEXT:    [[TMP11:%.*]] = mul nuw nsw i32 [[TMP7]], [[TMP4]]
; CHECK-NEXT:    [[TMP12:%.*]] = add i32 [[TMP10]], [[TMP11]]
; CHECK-NEXT:    [[TMP13:%.*]] = add i32 [[TMP12]], [[TMP8]]
; CHECK-NEXT:    [[TMP14:%.*]] = getelementptr inbounds [1024 x [4 x i32]], ptr addrspace(3) @scalar_alloca_ptr_with_vector_gep_offset_icmp_nullptr1.alloca, i32 0, i32 [[TMP13]]
; CHECK-NEXT:    [[GETELEMENTPTR0:%.*]] = getelementptr inbounds i8, ptr addrspace(3) [[TMP14]], <4 x i64> <i64 0, i64 1, i64 2, i64 3>
; CHECK-NEXT:    [[SELECT:%.*]] = select i1 [[COND]], <4 x ptr addrspace(3)> zeroinitializer, <4 x ptr addrspace(3)> [[GETELEMENTPTR0]]
; CHECK-NEXT:    [[ICMP:%.*]] = icmp eq <4 x ptr addrspace(3)> zeroinitializer, [[SELECT]]
; CHECK-NEXT:    store <4 x i1> [[ICMP]], ptr addrspace(1) [[OUT]], align 1
; CHECK-NEXT:    ret void
;
bb0:
  %alloca = alloca [4 x i32], align 4, addrspace(5)
  %getelementptr0 = getelementptr inbounds i8, ptr addrspace(5) %alloca, <4 x i64> <i64 0, i64 1, i64 2, i64 3>
  %select = select i1 %cond, <4 x ptr addrspace(5)> zeroinitializer, <4 x ptr addrspace(5)> %getelementptr0
  %icmp = icmp eq <4 x ptr addrspace(5)> zeroinitializer, %select
  store <4 x i1> %icmp, ptr addrspace(1) %out
  ret void
}

define amdgpu_kernel void @scalar_alloca_ptr_with_vector_gep_offset_phi_nullptr(i1 %cond, ptr addrspace(1) %out) {
; CHECK-LABEL: define amdgpu_kernel void @scalar_alloca_ptr_with_vector_gep_offset_phi_nullptr(
; CHECK-SAME: i1 [[COND:%.*]], ptr addrspace(1) [[OUT:%.*]]) {
; CHECK-NEXT:  [[BB0:.*]]:
; CHECK-NEXT:    [[TMP0:%.*]] = call noalias nonnull dereferenceable(64) ptr addrspace(4) @llvm.amdgcn.dispatch.ptr()
; CHECK-NEXT:    [[TMP1:%.*]] = getelementptr inbounds i32, ptr addrspace(4) [[TMP0]], i64 1
; CHECK-NEXT:    [[TMP2:%.*]] = load i32, ptr addrspace(4) [[TMP1]], align 4, !invariant.load [[META0]]
; CHECK-NEXT:    [[TMP3:%.*]] = getelementptr inbounds i32, ptr addrspace(4) [[TMP0]], i64 2
; CHECK-NEXT:    [[TMP4:%.*]] = load i32, ptr addrspace(4) [[TMP3]], align 4, !range [[RNG1]], !invariant.load [[META0]]
; CHECK-NEXT:    [[TMP5:%.*]] = lshr i32 [[TMP2]], 16
; CHECK-NEXT:    [[TMP6:%.*]] = call range(i32 0, 1024) i32 @llvm.amdgcn.workitem.id.x()
; CHECK-NEXT:    [[TMP7:%.*]] = call range(i32 0, 1024) i32 @llvm.amdgcn.workitem.id.y()
; CHECK-NEXT:    [[TMP8:%.*]] = call range(i32 0, 1024) i32 @llvm.amdgcn.workitem.id.z()
; CHECK-NEXT:    [[TMP9:%.*]] = mul nuw nsw i32 [[TMP5]], [[TMP4]]
; CHECK-NEXT:    [[TMP10:%.*]] = mul i32 [[TMP9]], [[TMP6]]
; CHECK-NEXT:    [[TMP11:%.*]] = mul nuw nsw i32 [[TMP7]], [[TMP4]]
; CHECK-NEXT:    [[TMP12:%.*]] = add i32 [[TMP10]], [[TMP11]]
; CHECK-NEXT:    [[TMP13:%.*]] = add i32 [[TMP12]], [[TMP8]]
; CHECK-NEXT:    [[TMP14:%.*]] = getelementptr inbounds [1024 x [4 x i32]], ptr addrspace(3) @scalar_alloca_ptr_with_vector_gep_offset_phi_nullptr.alloca, i32 0, i32 [[TMP13]]
; CHECK-NEXT:    br i1 [[COND]], label %[[BB1:.*]], label %[[BB2:.*]]
; CHECK:       [[BB1]]:
; CHECK-NEXT:    [[GETELEMENTPTR0:%.*]] = getelementptr inbounds i8, ptr addrspace(3) [[TMP14]], <4 x i64> <i64 0, i64 1, i64 2, i64 3>
; CHECK-NEXT:    br label %[[BB2]]
; CHECK:       [[BB2]]:
; CHECK-NEXT:    [[PHI:%.*]] = phi <4 x ptr addrspace(3)> [ [[GETELEMENTPTR0]], %[[BB1]] ], [ zeroinitializer, %[[BB0]] ]
; CHECK-NEXT:    [[EXTRACTELEMENT:%.*]] = extractelement <4 x ptr addrspace(3)> [[PHI]], i64 2
; CHECK-NEXT:    store i32 0, ptr addrspace(3) [[EXTRACTELEMENT]], align 4
; CHECK-NEXT:    ret void
;
bb0:
  %alloca = alloca [4 x i32], align 4, addrspace(5)
  br i1 %cond, label %bb1, label %bb2

bb1:
  %getelementptr0 = getelementptr inbounds i8, ptr addrspace(5) %alloca, <4 x i64> <i64 0, i64 1, i64 2, i64 3>
  br label %bb2

bb2:
  %phi = phi <4 x ptr addrspace(5)> [ %getelementptr0, %bb1 ], [ zeroinitializer, %bb0]
  %extractelement = extractelement <4 x ptr addrspace(5)> %phi, i64 2
  store i32 0, ptr addrspace(5) %extractelement
  ret void
}
;.
; CHECK: [[META0]] = !{}
; CHECK: [[RNG1]] = !{i32 0, i32 1025}
;.